Using ATR-FTIR spectra and convolutional neural networks for characterizing mixed plastic waste

Identification of Reaction Network Hypotheses for Complex Feedstocks from Spectroscopic Measurements with Minimal Human Intervention

In this work, we detail an automated reaction network hypothesis generation protocol for processes involving complex feedstocks where information about the species and reactions involved is unknown. Our methodology is process agnostic and can be utilized in any reactive process with spectroscopic measurements that provide information on the evolution of the components in the mixture. We decompose the mixture spectra to obtain spectroscopic signatures of the individual components and use a 1-D convolutional neural network to automatically identify functional groups indicated by them. We employ atom-atom mapping to automatically recover reaction rules that are applied on candidate molecules identified from chemistry databases through fingerprint similarity. The method is tested on synthetic data and on spectroscopic measurements of lab-scale batch hydrothermal liquefaction (HTL) of biomass to determine the accuracy of prediction across datasets of varying complexities. Our methodology is able to identify reaction network hypotheses containing reaction networks close to the ground truth in the case of synthetic data, and we are also able to recover candidate molecules and reaction networks close to the ones reported in the previous literature studies for biomass pyrolysis.

Artificial intelligence-empowered collection and characterization of microplastics: A review

Microplastics have been detected from water and soil systems extensively, with increasing evidence indicating their detrimental impacts on human and animal health. Concerns surrounding microplastic pollution have spurred the development of advanced collection and characterization methods for studying the size, abundance, distribution, chemical composition, and environmental impacts. This paper offers a comprehensive review of artificial intelligence (AI)-empowered technologies for the collection and characterization of microplastics. A framework is presented to streamline efforts in utilizing emerging robotics and machine learning technologies for collecting, processing, and characterizing microplastics. The review encompasses a range of AI technologies, delineating their principles, strengths, limitations, representative applications, and technology readiness levels, facilitating the selection of suitable AI technologies for mitigating microplastic pollution. New opportunities for future research and development on integrating robots and machine learning technologies are discussed to facilitate future efforts for mitigating microplastic pollution and advancing AI technologies.

Development of a machine learning model for systematics of Aspergillus section Nigri using synchrotron radiation-based fourier transform infrared spectroscopy

Aspergillus section Nigri (black aspergilli) fungi are economically important food spoilage agents. Some species in this section also produce harmful mycotoxins in food. However, it is remarkably difficult to identify this fungal group at the species level using morphological and chemical characteristics. The molecular approach for classification is preferable; however, it is time-consuming, making it inappropriate for rapid testing of large numbers of samples. To address this, we explored synchrotron radiation-based Fourier transform infrared microspectroscopy (SR-FTIR) as a rapid method for obtaining data suitable for species classification. SR-FTIR data were obtained from the mycelia/conidia of 22 black aspergilli species. The Convolutional Neural Network (CNN) approach, a supervised deep learning algorithm, was used with SR-FTIR data to classify black aspergilli at the species level. A subset of the data was used to train the CNN model, and the model classification performance was evaluated using the validation data subsets. The model demonstrated a 95.97% accuracy in species classification on the testing (blind) data subset. The technique presented herein could be an alternative method for identifying problematic black aspergilli in the food industry.

Convolutional Neural Networks Guided Raman Spectroscopy as a Process Analytical Technology (PAT) Tool for Monitoring and Simultaneous Prediction of Monoclonal Antibody Charge Variants

BACKGROUND

Charge related heterogeneities of monoclonal antibody (mAb) based therapeutic products are increasingly being considered as a critical quality attribute (CQA). They are typically estimated using analytical cation exchange chromatography (CEX), which is time consuming and not suitable for real time control. Raman spectroscopy coupled with artificial intelligence (AI) tools offers an opportunity for real time monitoring and control of charge variants.

OBJECTIVE

We present a process analytical technology (PAT) tool for on-line and real-time charge variant determination during process scale CEX based on Raman spectroscopy employing machine learning techniques.

METHOD

Raman spectra are collected from a reference library of samples with distribution of acidic, main, and basic species from 0-100% in a mAb concentration range of 0-20 g/L generated from process-scale CEX. The performance of different machine learning techniques for spectral processing is compared for predicting different charge variant species.

RESULT

A convolutional neural network (CNN) based model was successfully calibrated for quantification of acidic species, main species, basic species, and total protein concentration with R2 values of 0.94, 0.99, 0.96 and 0.99, respectively, and the Root Mean Squared Error (RMSE) of 0.1846, 0.1627, and 0.1029 g/L, respectively, and 0.2483 g/L for the total protein concentration.

CONCLUSION

We demonstrate that Raman spectroscopy combined with AI-ML frameworks can deliver rapid and accurate determination of product related impurities. This approach can be used for real time CEX pooling decisions in mAb production processes, thus enabling consistent charge variant profiles to be achieved.

Infrared Spectral Analysis for Prediction of Functional Groups Based on Feature-Aggregated Deep Learning

Infrared (IR) spectroscopy is a powerful and versatile tool for analyzing functional groups in organic compounds. A complex and time-consuming interpretation of massive unknown spectra usually requires knowledge of chemistry and spectroscopy. This paper presents a new deep learning method for transforming IR spectral features into intuitive imagelike feature maps and prediction of major functional groups. We obtained 8272 gas-phase IR spectra from the NIST Chemistry WebBook. Feature maps are constructed using the intrinsic correlation of spectral data, and prediction models are developed based on convolutional neural networks. Twenty-one major functional groups for each molecule are successfully identified using binary and multilabel models without expert guidance and feature selection. The multilabel classification model can produce all prediction results simultaneously for rapid characterization. Further analysis of the detailed substructures indicates that our model is capable of obtaining abundant structural information from IR spectra for a comprehensive investigation. The interpretation of our model reveals that the peaks of most interest are similar to those often considered by spectroscopists. In addition to demonstrating great potential for spectral identification, our method may contribute to the development of automated analyses in many fields.

Artificial intelligence-coupled plasmonic infrared sensor for detection of structural protein biomarkers in neurodegenerative diseases

Diagnosis of neurodegenerative disorders (NDDs) including Parkinson's disease and Alzheimer's disease is challenging owing to the lack of tools to detect preclinical biomarkers. The misfolding of proteins into oligomeric and fibrillar aggregates plays an important role in the development and progression of NDDs, thus underscoring the need for structural biomarker-based diagnostics. We developed an immunoassay-coupled nanoplasmonic infrared metasurface sensor that detects proteins linked to NDDs, such as alpha-synuclein, with specificity and differentiates the distinct structural species using their unique absorption signatures. We augmented the sensor with an artificial neural network enabling unprecedented quantitative prediction of oligomeric and fibrillar protein aggregates in their mixture. The microfluidic integrated sensor can retrieve time-resolved absorbance fingerprints in the presence of a complex biomatrix and is capable of multiplexing for the simultaneous monitoring of multiple pathology-associated biomarkers. Thus, our sensor is a promising candidate for the clinical diagnosis of NDDs, disease monitoring, and evaluation of novel therapies.

Framework for data-driven polymer characterization from infrared spectra

Automating infrared spectra interpretation in microplastic identification is of interest since most current methodologies are conducted manually or semi-automatically, which requires substantial processing time and presents a higher accuracy limited to single-polymer materials. Furthermore, when it comes to multicomponent or weathered polymeric materials commonly found in aquatic environments, identification usually becomes considerably depreciated as peaks shift and new signals are frequently observed, representing a significant deviation from reference spectral signatures. Therefore, this study aimed to develop a reference modeling framework for polymer identification through infrared spectra processing, addressing the limitations above. The case study selected for model development was polypropylene (PP) identification, as it is the second most abundant material in microplastics. Therefore, the database comprises 579 spectra with 52.3% containing PP to some degree. Different pretreatment and model parameters were evaluated for a more robust investigation, totaling 308 models, including multilayer perceptron and long-short-term memory architectures. The best model presented a test accuracy of 94.8% within the cross-validation standard deviation interval. Overall, the results achieved in this study indicate an opportunity to investigate the identification of other polymers following the same framework.

Spectral Classification of Large-Scale Blended (Micro)Plastics Using FT-IR Raw Spectra and Image-Based Machine Learning

Microplastics (MPs) are currently recognized as emerging pollutants; their identification and classification are therefore essential during their monitoring and management. In contrast to most studies based on small datasets and library searches, this study developed and compared four machine learning-based classifiers and two large-scale blended plastic datasets, where a 1D convolutional neural network (CNN), decision tree, and random forest (RF) were fed with raw spectral data from Fourier transform infrared spectroscopy, while a 2D CNN used the corresponding spectral images as the input. With an overall accuracy of 96.43% on a small dataset and 97.44% on a large dataset, the 1D CNN outperformed other models. The 1D CNN was the best at predicting environment samples, while the RF was the most robust with less spectral data. Overall, RF and 2D CNNs might be evaluated for plastic identification with fewer spectral data; however, 1D CNNs were thought to be the most effective with sufficient spectral data. Accordingly, an open-source MP spectroscopic analysis tool was developed to facilitate a quick and accurate analysis of existing MP samples.

Optical sensors and machine learning algorithms in sensor-based material flow characterization for mechanical recycling processes: A systematic literature review

Digital technologies hold enormous potential for improving the performance of future-generation sorting and processing plants; however, this potential remains largely untapped. Improved sensor-based material flow characterization (SBMC) methods could enable new sensor applications such as adaptive plant control, improved sensor-based sorting (SBS), and more far-reaching data utilizations along the value chain. This review aims to expedite research on SBMC by (i) providing a comprehensive overview of existing SBMC publications, (ii) summarizing existing SBMC methods, and (iii) identifying future research potentials in SBMC. By conducting a systematic literature search covering the period 2000 - 2021, we identified 198 peer-reviewed journal articles on SBMC applications based on optical sensors and machine learning algorithms for dry-mechanical recycling of non-hazardous waste. The review shows that SBMC has received increasing attention in recent years, with more than half of the reviewed publications published between 2019 and 2021. While applications were initially focused solely on SBS, the last decade has seen a trend toward new applications, including sensor-based material flow monitoring, quality control, and process monitoring/control. However, SBMC at the material flow and process level remains largely unexplored, and significant potential exists in upscaling investigations from laboratory to plant scale. Future research will benefit from a broader application of deep learning methods, increased use of low-cost sensors and new sensor technologies, and the use of data streams from existing SBS equipment. These advancements could significantly improve the performance of future-generation sorting and processing plants, keep more materials in closed loops, and help paving the way towards circular economy.